How should a conservation group write about science for a general reader?

Lead with the finding in one plain sentence, then give the conditions it depends on and how certain it is, and only then say what the organization thinks it means. Define technical terms where they first appear, keep numbers with their units, link the original study, and have the scientists who did the work review the summary before it publishes.

Conservation groups write for several readers at once: donors who care about a place, volunteers who work in it, journalists on deadline, agency staff, and other scientists. The temptation is to write two versions, one dense and one emotional, and let the emotional one carry the website. A general reader deserves the actual finding, in words they can follow.

This page covers writing and editing science content. It does not judge study design or statistics, which stay with your scientists. For the wider set of pages a conservation or research group needs, see our content overview for conservation and research groups.

Finding, meaning, and ask are separate sentences

Most overstated science writing comes from merging three things that should stay apart: what was measured, what the organization believes it means, and what it wants the reader to do. Each one has a different owner. One season is not a trend. A sentence that presents it as one puts the organization's credibility behind a claim its own researchers would not make.

The order matters as much as the separation. Readers who meet the ask first tend to read everything after it as persuasion, and the finding loses the authority it should lend to the request. Put the finding first, then the meaning, then the ask, even on a donation page, and keep each one in its own sentence so none of them borrows certainty from another.

Plain language without losing the science

Federal plain language guidance published on Digital.gov is a useful reference here. It says avoiding jargon does not mean leaving out necessary technical terms, and that when there is no other way to express an idea, writers should define their terms. It also warns that terms writers know well may be difficult or meaningless to their audience, and advises against long sections of definitions at the beginning or end of a document. For a conservation page, that means defining "riparian" in the sentence where it first appears, rather than in a glossary at the bottom, or replacing it with "streamside" when precision does not suffer.

Keep numbers with their units and their place: degrees, days, sites, a named reach of river. A rounded number with no unit is easy to repeat and hard to check. Charts follow the same rule. State the finding a chart shows in the sentence above it, so a reader who skips the image still gets it.

New data will sometimes complicate an older page. When it does, update the page, say what changed and when, and keep the earlier finding visible with its date rather than quietly deleting it. Readers who follow a place for years notice when a story changes without explanation.

Uncertainty words readers can follow

Scientists hedge for good reasons, and editors often strip the hedges out to make copy sound confident. Choose a small set of uncertainty words (for example "we measured," "suggests," "likely," "we do not yet know") and use them the same way on every page, so a reader learns what each one signals. Before publishing, hand the draft to the lead scientist with one question: "Is there a sentence here you would not say on the record?" Any sentence they mark either gets its hedge back or comes out.

Headlines, email subject lines, and social posts are where hedges disappear first, because they are written last and fastest. Hold them to the same standard as the body. If the body says warmer water "likely" reduces habitat, the headline cannot say it "is destroying" the river. A useful habit is to write the headline only after the scientist has approved the summary, and to send the headline back to them as well.

Sourcing, review, and AI drafts

Among the self-assessment questions Google publishes for site owners: is the content written or reviewed by an expert or enthusiast who demonstrably knows the topic well, and does it present clear sourcing, evidence of the expertise involved, and background about the author or the site? For a science page, the practical version is a link to the study or report, the names and roles of the researchers involved, and a reviewed-by line with a date. Drafting tools add a specific risk. A language model asked to make a summary readable may smooth away hedges or round numbers, which is exactly what the scientists will object to. If staff use one, the review step does not change: the lead scientist reads the final text against the report, line by line. The report itself should have its own landing page, as described in publishing research and reports so they can be cited.

A fictional example: the trout stream update

Consider a fictional river conservation group in Montana. Its spring appeal says the river is "dying" and that trout "face extinction." The monitoring team's report says something narrower: warmer summer water at four sites, two years of data, and a hypothesis about habitat. The revised page leads with the measured finding, gives the sites and years, states the hypothesis as a hypothesis, and then makes the ask. The writer should not have to invent what the data shows or how worried to sound, since the field team can say both in a ten-minute call and review the draft afterward.

Hudson South's four steps fit this kind of page in order. Authority Asset gathers the approved reports and findings. Hudson Twenty-One, the second step, records how the organization communicates across 21 dimensions so the writing sounds like it, while the scientists' own review stays the check on every claim; how Hudson Twenty-One works is described on its own page. Market Intelligence researches what readers ask about the river, and AEO Creation writes the answer-first pages. Keeping that register steady across many staff writers is its own question, covered in sounding consistent across many writers.

Check how your science pages read

Bring one research summary and the report behind it. We will spend a free 30-minute call comparing it with how peer groups explain similar work.

Sources

1. Avoid jargon. Digital.gov, U.S. General Services Administration. Read 2026-10-16.
2. Creating helpful, reliable, people-first content. Google Search Central. Read 2026-10-16.